298 BULLETIN : MUSEUM OF COMPARATIVE ZOOLOGY. 
to a network of lakelets and decaying ice-blocks. The frequent occur- 
rence of deltas with lobes at two or more levels favors this theory ; 
moreover, the extraordinary case of the Saxonville plain has already 
been explained as due to the wasting of an abandoned ice-block in Lake 
Cochituate. 
On the other hand, the remarkable way in which the many lobe alti- 
tudes fall into line, when plotted as in Plate 5, favors the idea that the 
discordance of levels, on the whole, is not irregular but systematic, — 
not so satisfactorily explained by isolated lakelets and wasting ice-blocks 
as by wide open lakes and an ice-front. which lay essentially east 
and west. 
Summary. 
My conclusions from this study of the sand plains of the Sudbury 
valley are these : — 
(a) In late glacial times the basin was occupied by a temporary ice- 
front lake. 
(6) This lake underwent successive lowerings of level, in a manner 
that may be deduced after the horizontal step scheme. At each stage, 
deltas were built. In some cases, changes of level occurred while the 
delta was under construction, — notably in the Saxonville sand plain, 
because that was built forward from a huge ice-block which occupied 
part of Lake Cochituate long after the main ice-sheet had left it. 
(c) Perhaps partly during the formation of the deltas—at any rate, 
after it —the whole region was tilted towards the south. By this tilt- 
ing apparent confusion of altitudes of the deltas was brought about ; but 
this discordance, when studied with proper accuracy and detail, is seen 
to be systematic. The tilt seems to be greatest in a due north-south 
direction, at the rate of about 7 feet per mile. 
Further detailed work, in determining the altitudes of the best lobes 
of a great number of sand plains in the extinct lake basins, ought to 
demonstrate the truth or incorrectness of the tilting hypothesis. 
In view of the fact that tilting to some degree is almost certain, from 
evidence in other parts of glaciated North America and along the New 
England coast, and that even a short field study of the seemingly dis- 
cordant deltas in the Sudbury valley goes far towards converting disorder 
into system, it appears that the grouping of sand plains in eastern 
Massachusetts may deserve even closer attention than it has usually 
received. 
